import pandas as pd
import matplotlib.pyplot as plt
import plotly.express as px
import seaborn as sns
import warnings
warnings.filterwarnings("ignore")
import plotly.io as pio
pio.renderers.default = 'notebook'

df = pd.read_csv('DajuBhaiMinimart.csv')
df

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 300 entries, 0 to 299
Data columns (total 16 columns):
 #   Column                 Non-Null Count  Dtype 
---  ------                 --------------  ----- 
 0   CustomerID             300 non-null    object
 1   CustomerName           300 non-null    object
 2   Quantity               300 non-null    int64 
 3   Price (NPR)            300 non-null    int64 
 4   Sales (NPR)            300 non-null    int64 
 5   Channel                300 non-null    object
 6   Ratings                300 non-null    int64 
 7   ProductCategory        300 non-null    object
 8   PaymentMethod          300 non-null    object
 9   TransactionDate        300 non-null    object
 10  StoreLocation          300 non-null    object
 11  DiscountApplied (NPR)  300 non-null    int64 
 12  LoyaltyMember          300 non-null    object
 13  CustomerContact        300 non-null    int64 
 14  Feedback               300 non-null    object
 15  Gender                 300 non-null    object
dtypes: int64(6), object(10)
memory usage: 37.6+ KB

df.isnull().sum()

CustomerID               0
CustomerName             0
Quantity                 0
Price (NPR)              0
Sales (NPR)              0
Channel                  0
Ratings                  0
ProductCategory          0
PaymentMethod            0
TransactionDate          0
StoreLocation            0
DiscountApplied (NPR)    0
LoyaltyMember            0
CustomerContact          0
Feedback                 0
Gender                   0
dtype: int64

# drop customer_id column
df.drop('CustomerID', axis=1, inplace=True)

#rename the column name
df = df.rename(columns={'Price (NPR)': 'Price'})
df = df.rename(columns={'Sales (NPR)': 'Sales'})
df = df.rename(columns={'DiscountApplied (NPR)': 'DiscountApplied'})

df

df.head()

df['ProductCategory'].unique()

array(['Electronics', 'Clothing', 'Personal Care', 'Groceries',
       'Household'], dtype=object)

df

# Calculate additional columns
df['TotalDiscount'] = df['DiscountApplied'] * df['Quantity']
df['CustomerFrequency'] = df.groupby('CustomerName')['CustomerName'].transform('count')
df['AverageRatingPerCustomer'] = df.groupby('CustomerName')['Ratings'].transform('mean')
df['TotalSalesPerCustomer'] = df.groupby('CustomerName')['Sales'].transform('sum')

# Extract date features before dropping TransactionDate
df['TransactionDate'] = pd.to_datetime(df['TransactionDate'])
df['IsWeekend'] = df['TransactionDate'].dt.dayofweek >= 5
df['IsWeekend'] = df['IsWeekend'].astype(int)

# Extract date features
df['TransactionDate'] = pd.to_datetime(df['TransactionDate'])
df['Year'] = df['TransactionDate'].dt.year
df['Month'] = df['TransactionDate'].dt.month
df['Day'] = df['TransactionDate'].dt.day
df['DayOfWeek'] = df['TransactionDate'].dt.dayofweek

# Map day of the week numbers to names
day_of_week_map = {0: 'Monday', 1: 'Tuesday', 2: 'Wednesday', 3: 'Thursday', 4: 'Friday', 5: 'Saturday', 6: 'Sunday'}
df['DayOfWeek'] = df['DayOfWeek'].map(day_of_week_map)

# Drop unnecessary columns
df.drop(columns=['CustomerName', 'CustomerContact', 'Feedback'], inplace=True)

df

# Group by Gender and count the number of visits
gender_distribution = df['Gender'].value_counts().reset_index()
gender_distribution.columns = ['Gender', 'Count']

# Create a pie chart
fig = px.pie(gender_distribution, 
             values='Count', 
             names='Gender', 
             title='Gender Distribution of Store Visitors')
fig.update_layout(width=800, height=600)

fig.show()

# visualizations based on Total Sales by Product Category plotly
total_sales_by_product_category = df.groupby('ProductCategory')['Sales'].sum().reset_index()
total_sales_by_product_category = total_sales_by_product_category.sort_values(by='Sales', ascending=False)
#bar
fig = px.bar(total_sales_by_product_category, x='ProductCategory', y='Sales',color='ProductCategory', title='Total Sales by Product Category')
fig.update_layout(xaxis_title="Product Category", yaxis_title="Total Sales")
fig.show()

# visualization based average rating by product category
average_rating_by_product_category = df.groupby('ProductCategory')['Ratings'].mean().reset_index()
average_rating_by_product_category = average_rating_by_product_category.sort_values(by='Ratings', ascending=False)

#pie chart
fig = px.pie(average_rating_by_product_category, names='ProductCategory', values='Ratings', title='Average Rating by Product Category')
fig.show()

#total Sales by payment method
total_sales_by_payment_method = df.groupby('PaymentMethod')['Sales'].sum().reset_index()
total_sales_by_payment_method = total_sales_by_payment_method.sort_values(by='Sales', ascending=False)

# histogram
fig = px.histogram(df, x='PaymentMethod', y='Sales', color='PaymentMethod', title='Total Sales by Payment Method', histfunc='sum')
fig.update_layout(xaxis_title="Payment Method", yaxis_title="Total Sales")
fig.show()

#total sales by channel
total_sales_by_channel = df.groupby('Channel')['Sales'].sum().reset_index()
total_sales_by_channel = total_sales_by_channel.sort_values(by='Sales', ascending=False)

#bar plotly
fig = px.bar(total_sales_by_channel, x='Channel', y='Sales',color='Sales',color_continuous_scale='temps', title='TotalSales by Channel')
fig.update_layout(xaxis_title="Channel", yaxis_title="Total Sales")
fig.show()

#total sales by day of the week
total_sales_by_day = df.groupby('DayOfWeek')['Sales'].sum().reset_index()
days_order = ['Sunday', 'Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday']
total_sales_by_day['DayOfWeek'] = pd.Categorical(total_sales_by_day['DayOfWeek'], categories=days_order, ordered=True)
total_sales_by_day = total_sales_by_day.sort_values('DayOfWeek')

fig = px.bar(total_sales_by_day, x='DayOfWeek', y='Sales', title='Total Sales by Day of the Week', color='Sales', color_continuous_scale='temps')
fig.update_layout(xaxis_title="Day of the Week", yaxis_title="Total Sales (NPR)")
fig.show()

#visualizations based on Total Sales by Month
total_sales_by_month = df.groupby('Month')['Sales'].sum().reset_index()
fig = px.scatter(total_sales_by_month, x='Month', y='Sales', title='Total Sales by Month',color='Sales',color_continuous_scale='matter')
fig.show()

# Calculate average discount applied by product category
average_discount_by_category = df.groupby('ProductCategory')['DiscountApplied'].mean().reset_index()

# Create a line graph
fig = px.line(average_discount_by_category, x='ProductCategory', y='DiscountApplied', title='Average Discount Applied by Product Category')
fig.update_layout(xaxis_title="Product Category", yaxis_title="Average Discount Applied")
fig.show()

#total sales by gender
sales_by_gender = df.groupby('Gender')['Sales'].sum().reset_index()
#pie
fig = px.pie(sales_by_gender, values='Sales', names='Gender', title='Total Sales by Gender',color='Gender')
fig.show()

# Visualize the total quantity sold by product category
quantity_by_category = df.groupby('ProductCategory')['Quantity'].sum().reset_index()

fig = px.bar(quantity_by_category, 
                 x='ProductCategory', 
                 y='Quantity', 
                 title='Total Quantity Sold by Product Category',
                 color='Quantity',
                 color_continuous_scale='matter',
)
# Update the layout to make the plot bigger
fig.update_layout(width=800, height=600)

fig.show()

#loyality member 
df['LoyaltyMember'] = df['LoyaltyMember'].map({
    'Yes': '1',
    'No': '0'
})

#product Category mapping
df['ProductCategory'] = df['ProductCategory'].map({
    'Electronics': '0',
    'Clothing': '1',
    'Groceries': '2',
    'Personal Care': '3',
    'Household': '4'
})

#mapping channel name
df['Channel'] = df['Channel'].map({
    'Website': '0',
    'Market Visit': '1'
})
df['Channel'] = pd.to_numeric(df['Channel'], errors='coerce')

# payment method mapping
df['PaymentMethod'] = df['PaymentMethod'].map({
    'Card': '0',
    'Mobile Payment': '1',
    'Cash': '2'
})
df.head()

# Ratings Category (Low, Medium, High)
def categorize_rating(Ratings):
    if Ratings <= 2.5:
        return 'Low'
    elif Ratings <= 4.0:
        return 'Medium'
    else:
        return 'High'

df['Rating_Category'] = df['Ratings'].apply(categorize_rating)
df.head()

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 300 entries, 0 to 299
Data columns (total 22 columns):
 #   Column                    Non-Null Count  Dtype         
---  ------                    --------------  -----         
 0   Quantity                  300 non-null    int64         
 1   Price                     300 non-null    int64         
 2   Sales                     300 non-null    int64         
 3   Channel                   300 non-null    int64         
 4   Ratings                   300 non-null    int64         
 5   ProductCategory           300 non-null    object        
 6   PaymentMethod             300 non-null    object        
 7   TransactionDate           300 non-null    datetime64[ns]
 8   StoreLocation             300 non-null    object        
 9   DiscountApplied           300 non-null    int64         
 10  LoyaltyMember             300 non-null    object        
 11  Gender                    300 non-null    object        
 12  TotalDiscount             300 non-null    int64         
 13  CustomerFrequency         300 non-null    int64         
 14  AverageRatingPerCustomer  300 non-null    float64       
 15  TotalSalesPerCustomer     300 non-null    int64         
 16  IsWeekend                 300 non-null    int64         
 17  Year                      300 non-null    int32         
 18  Month                     300 non-null    int32         
 19  Day                       300 non-null    int32         
 20  DayOfWeek                 300 non-null    object        
 21  Rating_Category           300 non-null    object        
dtypes: datetime64[ns](1), float64(1), int32(3), int64(10), object(7)
memory usage: 48.2+ KB

#data to numeric to columns
df['ProductCategory'] = pd.to_numeric(df['ProductCategory'], errors='coerce')
df['PaymentMethod'] = pd.to_numeric(df['PaymentMethod'], errors='coerce')
df['LoyalityMember'] = pd.to_numeric(df['LoyaltyMember'], errors='coerce')
df.head()

df.info()
df['Channel'].unique()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 300 entries, 0 to 299
Data columns (total 23 columns):
 #   Column                    Non-Null Count  Dtype         
---  ------                    --------------  -----         
 0   Quantity                  300 non-null    int64         
 1   Price                     300 non-null    int64         
 2   Sales                     300 non-null    int64         
 3   Channel                   300 non-null    int64         
 4   Ratings                   300 non-null    int64         
 5   ProductCategory           300 non-null    int64         
 6   PaymentMethod             300 non-null    int64         
 7   TransactionDate           300 non-null    datetime64[ns]
 8   StoreLocation             300 non-null    object        
 9   DiscountApplied           300 non-null    int64         
 10  LoyaltyMember             300 non-null    object        
 11  Gender                    300 non-null    object        
 12  TotalDiscount             300 non-null    int64         
 13  CustomerFrequency         300 non-null    int64         
 14  AverageRatingPerCustomer  300 non-null    float64       
 15  TotalSalesPerCustomer     300 non-null    int64         
 16  IsWeekend                 300 non-null    int64         
 17  Year                      300 non-null    int32         
 18  Month                     300 non-null    int32         
 19  Day                       300 non-null    int32         
 20  DayOfWeek                 300 non-null    object        
 21  Rating_Category           300 non-null    object        
 22  LoyalityMember            300 non-null    int64         
dtypes: datetime64[ns](1), float64(1), int32(3), int64(13), object(5)
memory usage: 50.5+ KB

array([0, 1])

df.describe()

corr = df.corr(numeric_only=1)
corr

# Heatmap
plt.figure(figsize=(12, 8))
sns.heatmap(corr, annot=True, fmt='.2f', cmap='magma', linewidths=0.5)
plt.title('Correlation Heatmap')
plt.show()

from sklearn.model_selection import train_test_split, cross_val_score
from sklearn.ensemble import GradientBoostingRegressor, RandomForestRegressor
from sklearn.metrics import mean_absolute_error,r2_score

Features= df[['Quantity', 'Price','ProductCategory', 'PaymentMethod', 'LoyaltyMember']]
Target = df['Sales']

X_train, X_test, y_train, y_test = train_test_split(Features, Target, test_size=0.2, random_state=42)

#gradient boosting regressor
model = GradientBoostingRegressor(random_state=42)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

mse = mean_absolute_error(y_test, y_pred)
r2_reg= r2_score(y_test, y_pred)

print(f'Mean Absolute Error: {mse}')
print(f'R-squared: {r2_reg}')
#to view the actual and predicted values
for actual, predicted in zip(y_test, y_pred):
    print(f"Actual: {actual}, Predicted: {int(predicted)}")

Mean Absolute Error: 0.12313124521990503
R-squared: 0.8302825355373844
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0

# Visualization of predicted vs actual costs
fig_cost = px.scatter(x=y_test, y=y_pred, labels={'x': 'Actual Cost', 'y': 'Predicted Cost'}, title='Actual vs Predicted Cost')
fig_cost.add_shape(
    type="line", line=dict(dash='dash'),
    x0=y_test.min(), y0=y_test.min(),
    x1=y_test.max(), y1=y_test.max()
)
fig_cost.update_layout(paper_bgcolor="white")
fig_cost.show()

#hyperparameter tuning and cross validation for gradientboostingRegressor
gbr = GradientBoostingRegressor(
    random_state=42, 
    n_estimators=200,         
    max_depth=3,           
    min_samples_split=2,      
    min_samples_leaf=2, 
    learning_rate=0.1,      
)

gbr.fit(X_train, y_train)

y_pred_reg = gbr.predict(X_test)

cv_scores_r2_reg = cross_val_score(gbr, Features, Target, cv=5, scoring='r2')
cv_scores_mae = cross_val_score(gbr, Features, Target, cv=5, scoring='neg_mean_absolute_error')
cv_scores_mae = -cv_scores_mae 
print("Mean R2 score:", cv_scores_r2_reg.mean())
print("Mean MAE score:", cv_scores_mae.mean())
#to view the actual and predicted values
for actual, predicted in zip(y_test, y_pred_reg):
    print(f"Actual: {actual}, Predicted: {int(predicted)}")

Mean R2 score: 0.9926229878124989
Mean MAE score: 348.9119389870635
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 1, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 1
Actual: 0, Predicted: 0
Actual: 0, Predicted: 0
Actual: 1, Predicted: 0

# Visualization of predicted vs actual costs after hyperparameter tuning
fig_cost = px.scatter(x=y_test, y=y_pred_reg, labels={'x': 'Actual Cost', 'y': 'Predicted Cost'}, title='Actual vs Predicted Cost')
fig_cost.add_shape(
    type="line", line=dict(dash='dash'),
    x0=y_test.min(), y0=y_test.min(),
    x1=y_test.max(), y1=y_test.max()
)
fig_cost.update_layout(paper_bgcolor="white")
fig_cost.show()

#save the model
import joblib
joblib.dump(gbr, 'sales_prediction_model.pkl')

['sales_prediction_model.pkl']

# for classification of the loyalty member
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

Features_clf = df[['Quantity','Sales', 'ProductCategory', 'PaymentMethod', 'DiscountApplied', 'TotalDiscount', 'CustomerFrequency', 'AverageRatingPerCustomer', 'TotalSalesPerCustomer']]
Target_clf = df['LoyaltyMember']

X_train, X_test, y_train, y_test = train_test_split(Features_clf, Target_clf, test_size=0.2, random_state=42)

model_clf = RandomForestClassifier()

model_clf.fit(X_train, y_train)
y_pred_clf = model_clf.predict(X_test)

accuracy = accuracy_score(y_test, y_pred_clf)
print(f'Accuracy: {accuracy}')
print(classification_report(y_test, y_pred_clf))

Accuracy: 0.8833333333333333
              precision    recall  f1-score   support

           0       0.97      0.83      0.89        35
           1       0.80      0.96      0.87        25

    accuracy                           0.88        60
   macro avg       0.88      0.89      0.88        60
weighted avg       0.90      0.88      0.88        60

# confusion matrix by plotly
import plotly.graph_objects as go
from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred_clf)

# Create a confusion matrix heatmap
fig = go.Figure(data=go.Heatmap(
    z=cm,
    x=['Predicted Negative', 'Predicted Positive'],
    y=['Actual Negative', 'Actual Positive'],
    hoverongaps=False,
    colorscale='Viridis'))

fig.update_layout(
    title='Confusion Matrix',
    xaxis_title='Predicted Label',
    yaxis_title='True Label'
)

fig.show()

# random forest classifier
from sklearn.tree import plot_tree

num_estimators = len(model_clf.estimators_)
print(f'The model has {num_estimators} estimators.')

plt.figure(figsize=(20, 10))
plot_tree(model_clf.estimators_[0], filled=True, feature_names=Features_clf.columns, class_names=[str(cls) for cls in model_clf.classes_])
plt.show()

The model has 100 estimators.

#hyperparameter tuning and cross validation for RandomForestClassifier
rfc = RandomForestClassifier(
    random_state=42, 
    n_estimators=300,         
    max_depth=3,           
    min_samples_split=2,      
    min_samples_leaf=2,       
)

rfc.fit(X_train, y_train)
y_pred_rfc = rfc.predict(X_test)

cv_scores_rfc = cross_val_score(rfc, Features_clf, Target_clf, cv=5, scoring='accuracy')
print("Cross-validation accuracy scores for Random Forest Classifier:", cv_scores_rfc)
print("Mean accuracy score:", cv_scores_rfc.mean())

Cross-validation accuracy scores for Random Forest Classifier: [0.93333333 0.7        0.73333333 0.78333333 0.8       ]
Mean accuracy score: 0.79

#to view the actual and predicted values
for actual, predicted in zip(y_test, y_pred_reg):
    print(f"Actual: {actual}, Predicted: {int(predicted)}")

Actual: 0, Predicted: 978
Actual: 1, Predicted: 22402
Actual: 1, Predicted: 2895
Actual: 1, Predicted: 10013
Actual: 0, Predicted: 10082
Actual: 1, Predicted: 7024
Actual: 1, Predicted: 16117
Actual: 1, Predicted: 21932
Actual: 0, Predicted: 7211
Actual: 0, Predicted: 919
Actual: 0, Predicted: 9603
Actual: 1, Predicted: 23343
Actual: 0, Predicted: 2786
Actual: 0, Predicted: 3865
Actual: 0, Predicted: 5421
Actual: 0, Predicted: 4372
Actual: 1, Predicted: 4517
Actual: 0, Predicted: 2125
Actual: 0, Predicted: 2261
Actual: 0, Predicted: 1195
Actual: 1, Predicted: 2196
Actual: 0, Predicted: 11410
Actual: 0, Predicted: 759
Actual: 1, Predicted: 3339
Actual: 0, Predicted: 656
Actual: 1, Predicted: 860
Actual: 0, Predicted: 4539
Actual: 0, Predicted: 5607
Actual: 1, Predicted: 5883
Actual: 0, Predicted: 2907
Actual: 1, Predicted: 14658
Actual: 1, Predicted: 7766
Actual: 1, Predicted: 3138
Actual: 1, Predicted: 1527
Actual: 1, Predicted: 8027
Actual: 0, Predicted: 2877
Actual: 0, Predicted: 3717
Actual: 0, Predicted: 20326
Actual: 0, Predicted: 9513
Actual: 1, Predicted: 3037
Actual: 0, Predicted: 515
Actual: 1, Predicted: 21791
Actual: 0, Predicted: 10407
Actual: 1, Predicted: 2360
Actual: 0, Predicted: 1641
Actual: 0, Predicted: 1439
Actual: 0, Predicted: 706
Actual: 1, Predicted: 3131
Actual: 0, Predicted: 2502
Actual: 1, Predicted: 3677
Actual: 0, Predicted: 13820
Actual: 0, Predicted: 4010
Actual: 1, Predicted: 10658
Actual: 0, Predicted: 8420
Actual: 0, Predicted: 2593
Actual: 0, Predicted: 3445
Actual: 1, Predicted: 1822
Actual: 0, Predicted: 7431
Actual: 0, Predicted: 5725
Actual: 1, Predicted: 2822

	CustomerID	CustomerName	Quantity	Price (NPR)	Sales (NPR)	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	StoreLocation	DiscountApplied (NPR)	LoyaltyMember	CustomerContact	Feedback	Gender
0	CUST001	Bishal Gurung	11	1303	17199	Website	4	Electronics	Card	12/6/2023 5:13	-	60	Yes	9819894305	Quick delivery	Male
1	CUST002	Sita Tamang	10	610	7246	Market Visit	4	Clothing	Mobile Payment	12/22/2023 12:54	Pokhara	240	Yes	9823584905	Satisfied with quality	Female
2	CUST003	Sarita Lama	7	1812	12684	Market Visit	3	Electronics	Mobile Payment	10/13/2024 1:03	Lalitpur	200	No	9823088823	Friendly staff	Female
3	CUST004	Ram Shrestha	3	1070	3210	Website	1	Electronics	Mobile Payment	6/5/2024 6:08	-	100	No	9888996168	Quick delivery	Male
4	CUST005	Sita Tamang	10	1209	14362	Market Visit	5	Personal Care	Card	10/26/2024 16:28	Bhaktapur	60	Yes	9874939195	Prices a bit high	Female
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
295	CUST296	Kamal Sharma	6	919	5514	Website	2	Personal Care	Mobile Payment	4/23/2024 5:18	-	50	No	9835949686	Satisfied with quality	Male
296	CUST297	Asha Rai	7	685	4795	Market Visit	3	Groceries	Mobile Payment	10/5/2024 5:05	Lalitpur	0	No	9869059335	Satisfied with quality	Male
297	CUST298	Kamal Sharma	8	342	2736	Market Visit	1	Electronics	Card	12/1/2023 8:10	Pokhara	200	No	9866858364	Prices a bit high	Male
298	CUST299	Gurax Gurung	11	217	2864	Market Visit	4	Household	Cash	1/21/2024 2:00	Pokhara	60	Yes	9816458335	Friendly staff	Female
299	CUST300	Sarita Lama	9	1644	17360	Market Visit	3	Clothing	Cash	11/12/2023 14:47	Lalitpur	0	Yes	9893314124	Prices a bit high	Female

	CustomerName	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	StoreLocation	DiscountApplied	LoyaltyMember	CustomerContact	Feedback	Gender
0	Bishal Gurung	11	1303	17199	Website	4	Electronics	Card	12/6/2023 5:13	-	60	Yes	9819894305	Quick delivery	Male
1	Sita Tamang	10	610	7246	Market Visit	4	Clothing	Mobile Payment	12/22/2023 12:54	Pokhara	240	Yes	9823584905	Satisfied with quality	Female
2	Sarita Lama	7	1812	12684	Market Visit	3	Electronics	Mobile Payment	10/13/2024 1:03	Lalitpur	200	No	9823088823	Friendly staff	Female
3	Ram Shrestha	3	1070	3210	Website	1	Electronics	Mobile Payment	6/5/2024 6:08	-	100	No	9888996168	Quick delivery	Male
4	Sita Tamang	10	1209	14362	Market Visit	5	Personal Care	Card	10/26/2024 16:28	Bhaktapur	60	Yes	9874939195	Prices a bit high	Female
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
295	Kamal Sharma	6	919	5514	Website	2	Personal Care	Mobile Payment	4/23/2024 5:18	-	50	No	9835949686	Satisfied with quality	Male
296	Asha Rai	7	685	4795	Market Visit	3	Groceries	Mobile Payment	10/5/2024 5:05	Lalitpur	0	No	9869059335	Satisfied with quality	Male
297	Kamal Sharma	8	342	2736	Market Visit	1	Electronics	Card	12/1/2023 8:10	Pokhara	200	No	9866858364	Prices a bit high	Male
298	Gurax Gurung	11	217	2864	Market Visit	4	Household	Cash	1/21/2024 2:00	Pokhara	60	Yes	9816458335	Friendly staff	Female
299	Sarita Lama	9	1644	17360	Market Visit	3	Clothing	Cash	11/12/2023 14:47	Lalitpur	0	Yes	9893314124	Prices a bit high	Female

	CustomerName	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	StoreLocation	DiscountApplied	LoyaltyMember	CustomerContact	Feedback	Gender
0	Bishal Gurung	11	1303	17199	Website	4	Electronics	Card	12/6/2023 5:13	-	60	Yes	9819894305	Quick delivery	Male
1	Sita Tamang	10	610	7246	Market Visit	4	Clothing	Mobile Payment	12/22/2023 12:54	Pokhara	240	Yes	9823584905	Satisfied with quality	Female
2	Sarita Lama	7	1812	12684	Market Visit	3	Electronics	Mobile Payment	10/13/2024 1:03	Lalitpur	200	No	9823088823	Friendly staff	Female
3	Ram Shrestha	3	1070	3210	Website	1	Electronics	Mobile Payment	6/5/2024 6:08	-	100	No	9888996168	Quick delivery	Male
4	Sita Tamang	10	1209	14362	Market Visit	5	Personal Care	Card	10/26/2024 16:28	Bhaktapur	60	Yes	9874939195	Prices a bit high	Female
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
295	Kamal Sharma	6	919	5514	Website	2	Personal Care	Mobile Payment	4/23/2024 5:18	-	50	No	9835949686	Satisfied with quality	Male
296	Asha Rai	7	685	4795	Market Visit	3	Groceries	Mobile Payment	10/5/2024 5:05	Lalitpur	0	No	9869059335	Satisfied with quality	Male
297	Kamal Sharma	8	342	2736	Market Visit	1	Electronics	Card	12/1/2023 8:10	Pokhara	200	No	9866858364	Prices a bit high	Male
298	Gurax Gurung	11	217	2864	Market Visit	4	Household	Cash	1/21/2024 2:00	Pokhara	60	Yes	9816458335	Friendly staff	Female
299	Sarita Lama	9	1644	17360	Market Visit	3	Clothing	Cash	11/12/2023 14:47	Lalitpur	0	Yes	9893314124	Prices a bit high	Female

	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	StoreLocation	DiscountApplied	...	Gender	TotalDiscount	CustomerFrequency	AverageRatingPerCustomer	TotalSalesPerCustomer	IsWeekend	Year	Month	Day	DayOfWeek
0	11	1303	17199	Website	4	Electronics	Card	2023-12-06 05:13:00	-	60	...	Male	660	49	3.693878	371058	0	2023	12	6	Wednesday
1	10	610	7246	Market Visit	4	Clothing	Mobile Payment	2023-12-22 12:54:00	Pokhara	240	...	Female	2400	47	3.361702	341186	0	2023	12	22	Friday
2	7	1812	12684	Market Visit	3	Electronics	Mobile Payment	2024-10-13 01:03:00	Lalitpur	200	...	Female	1400	42	3.380952	278411	1	2024	10	13	Sunday
3	3	1070	3210	Website	1	Electronics	Mobile Payment	2024-06-05 06:08:00	-	100	...	Male	300	43	3.372093	211555	0	2024	6	5	Wednesday
4	10	1209	14362	Market Visit	5	Personal Care	Card	2024-10-26 16:28:00	Bhaktapur	60	...	Female	600	47	3.361702	341186	1	2024	10	26	Saturday
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
295	6	919	5514	Website	2	Personal Care	Mobile Payment	2024-04-23 05:18:00	-	50	...	Male	300	47	3.531915	383490	0	2024	4	23	Tuesday
296	7	685	4795	Market Visit	3	Groceries	Mobile Payment	2024-10-05 05:05:00	Lalitpur	0	...	Male	0	36	3.361111	243614	1	2024	10	5	Saturday
297	8	342	2736	Market Visit	1	Electronics	Card	2023-12-01 08:10:00	Pokhara	200	...	Male	1600	47	3.531915	383490	0	2023	12	1	Friday
298	11	217	2864	Market Visit	4	Household	Cash	2024-01-21 02:00:00	Pokhara	60	...	Female	660	1	4.000000	2864	1	2024	1	21	Sunday
299	9	1644	17360	Market Visit	3	Clothing	Cash	2023-11-12 14:47:00	Lalitpur	0	...	Female	0	42	3.380952	278411	1	2023	11	12	Sunday

	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	StoreLocation	DiscountApplied	...	Gender	TotalDiscount	CustomerFrequency	AverageRatingPerCustomer	TotalSalesPerCustomer	IsWeekend	Year	Month	Day	DayOfWeek
0	11	1303	17199	0	4	0	0	2023-12-06 05:13:00	-	60	...	Male	660	49	3.693878	371058	0	2023	12	6	Wednesday
1	10	610	7246	1	4	1	1	2023-12-22 12:54:00	Pokhara	240	...	Female	2400	47	3.361702	341186	0	2023	12	22	Friday
2	7	1812	12684	1	3	0	1	2024-10-13 01:03:00	Lalitpur	200	...	Female	1400	42	3.380952	278411	1	2024	10	13	Sunday
3	3	1070	3210	0	1	0	1	2024-06-05 06:08:00	-	100	...	Male	300	43	3.372093	211555	0	2024	6	5	Wednesday
4	10	1209	14362	1	5	3	0	2024-10-26 16:28:00	Bhaktapur	60	...	Female	600	47	3.361702	341186	1	2024	10	26	Saturday

Data Cleaning¶

Data Visualization¶

Feature Engineering¶

Model Training¶

Hyperparameter tuning and cross validation for Gradient Boosting Regressor¶

Hyperparameter tuning and cross validation for RandomForestClassifier¶

	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	TransactionDate	DiscountApplied	TotalDiscount	CustomerFrequency	AverageRatingPerCustomer	TotalSalesPerCustomer	IsWeekend	Year	Month	Day	LoyalityMember
count	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000	300.000000
mean	6.126667	1046.113333	6848.303333	0.470000	3.473333	1.946667	0.946667	2024-05-14 01:22:03.600000	98.366667	603.300000	41.733333	3.473333	291138.376667	0.296667	2023.853333	6.673333	15.666667	0.490000
min	1.000000	107.000000	145.000000	0.000000	1.000000	0.000000	0.000000	2023-11-10 23:24:00	0.000000	0.000000	1.000000	1.000000	1071.000000	0.000000	2023.000000	1.000000	1.000000	0.000000
25%	4.000000	532.000000	2413.250000	0.000000	2.000000	1.000000	0.000000	2024-02-07 20:28:45	50.000000	87.500000	42.000000	3.361702	211555.000000	0.000000	2024.000000	4.000000	7.750000	0.000000
50%	6.000000	1086.000000	5165.500000	0.000000	4.000000	2.000000	1.000000	2024-05-19 12:17:00	100.000000	480.000000	43.000000	3.380952	278411.000000	0.000000	2024.000000	7.000000	16.000000	0.000000
75%	9.000000	1524.250000	10034.000000	1.000000	5.000000	3.000000	2.000000	2024-08-11 11:19:00	120.000000	960.000000	47.000000	3.535714	371058.000000	1.000000	2024.000000	10.000000	24.000000	1.000000
max	11.000000	1998.000000	25819.000000	1.000000	5.000000	4.000000	2.000000	2024-11-04 18:03:00	240.000000	2640.000000	49.000000	5.000000	383490.000000	1.000000	2024.000000	12.000000	31.000000	1.000000
std	2.997874	554.192076	5634.298122	0.499933	1.332369	1.500442	0.824270	NaN	79.662744	593.775497	9.144815	0.235653	83860.974326	0.457552	0.354364	3.465061	9.047379	0.500735

	Quantity	Price	Sales	Channel	Ratings	ProductCategory	PaymentMethod	DiscountApplied	TotalDiscount	CustomerFrequency	AverageRatingPerCustomer	TotalSalesPerCustomer	IsWeekend	Year	Month	Day	LoyalityMember
Quantity	1.000000	0.012748	0.668159	0.058332	-0.079534	0.029761	0.036580	0.002690	0.482048	0.032955	-0.058428	0.084254	0.001772	-0.010788	-0.024658	-0.051461	0.183539
Price	0.012748	1.000000	0.647220	-0.044857	0.068485	-0.012232	-0.035840	-0.091466	-0.086800	0.027917	0.109680	0.092431	0.022302	-0.021935	0.045250	-0.073137	0.003668
Sales	0.668159	0.647220	1.000000	-0.018217	0.038459	0.004102	0.007956	-0.073233	0.233778	0.103906	0.042509	0.166571	0.035740	-0.033195	0.020542	-0.113691	0.173722
Channel	0.058332	-0.044857	-0.018217	1.000000	0.026411	0.100407	-0.028244	0.169659	0.133675	0.009217	0.011350	0.007948	0.104833	-0.043798	0.061897	0.020704	-0.001202
Ratings	-0.079534	0.068485	0.038459	0.026411	1.000000	0.005978	0.044381	0.054888	-0.027304	-0.008820	0.176868	0.019365	-0.011667	-0.015395	-0.032319	-0.055120	0.307897
ProductCategory	0.029761	-0.012232	0.004102	0.100407	0.005978	1.000000	-0.018533	0.026690	0.054105	-0.002746	-0.063846	-0.007227	0.076711	0.054431	-0.022017	-0.019053	0.030448
PaymentMethod	0.036580	-0.035840	0.007956	-0.028244	0.044381	-0.018533	1.000000	0.018533	-0.009548	-0.079983	0.087827	-0.029193	0.033225	-0.015419	-0.132586	0.063982	0.071631
DiscountApplied	0.002690	-0.091466	-0.073233	0.169659	0.054888	0.026690	0.018533	1.000000	0.787155	0.041269	0.027291	0.039173	0.060134	0.049538	-0.106865	0.012745	0.150925
TotalDiscount	0.482048	-0.086800	0.233778	0.133675	-0.027304	0.054105	-0.009548	0.787155	1.000000	0.069091	0.027262	0.091305	0.018420	0.034892	-0.088651	-0.000299	0.182845
CustomerFrequency	0.032955	0.027917	0.103906	0.009217	-0.008820	-0.002746	-0.079983	0.041269	0.069091	1.000000	-0.049870	0.867935	0.002984	-0.058552	0.106904	-0.011709	0.041777
AverageRatingPerCustomer	-0.058428	0.109680	0.042509	0.011350	0.176868	-0.063846	0.087827	0.027291	0.027262	-0.049870	1.000000	0.109487	-0.030067	0.015437	-0.013972	0.086413	0.048151
TotalSalesPerCustomer	0.084254	0.092431	0.166571	0.007948	0.019365	-0.007227	-0.029193	0.039173	0.091305	0.867935	0.109487	1.000000	-0.036349	-0.017618	0.049690	0.008862	0.050110
IsWeekend	0.001772	0.022302	0.035740	0.104833	-0.011667	0.076711	0.033225	0.060134	0.018420	0.002984	-0.030067	-0.036349	1.000000	-0.060781	0.002264	-0.025315	0.034888
Year	-0.010788	-0.021935	-0.033195	-0.043798	-0.015395	0.054431	-0.015419	0.049538	0.034892	-0.058552	0.015437	-0.017618	-0.060781	1.000000	-0.616585	-0.065372	0.010555
Month	-0.024658	0.045250	0.020542	0.061897	-0.032319	-0.022017	-0.132586	-0.106865	-0.088651	0.106904	-0.013972	0.049690	0.002264	-0.616585	1.000000	0.048896	-0.113688
Day	-0.051461	-0.073137	-0.113691	0.020704	-0.055120	-0.019053	0.063982	0.012745	-0.000299	-0.011709	0.086413	0.008862	-0.025315	-0.065372	0.048896	1.000000	-0.033221
LoyalityMember	0.183539	0.003668	0.173722	-0.001202	0.307897	0.030448	0.071631	0.150925	0.182845	0.041777	0.048151	0.050110	0.034888	0.010555	-0.113688	-0.033221	1.000000